AIM Lab LogoThe AIM Lab (AI-Driven Behavioral Health and Immersive Technologies Innovation Lab) focuses on advancing behavioral health through the integration of artificial intelligence and immersive technologies. Our work develops scalable, data-driven solutions to improve mental health and substance use outcomes, with a particular emphasis on crisis response, communication, and early risk identification. By leveraging machine learning, natural language processing, and virtual reality platforms, we design next-generation interventions that enhance real-world decision-making and support practitioners in high-stakes environments.

Through interdisciplinary collaboration across behavioral health, public safety, and technology, the AIM Lab translates research into actionable tools, including AI-driven decision-support systems and immersive training platforms. Our goal is to bridge the gap between innovation and practice by delivering culturally responsive, human-centered solutions that can be implemented at scale to improve outcomes across diverse populations and community settings.

Mission

To advance behavioral health outcomes through the development and application of artificial intelligence and immersive technologies, creating scalable, culturally responsive solutions that improve assessment, decision-making, and intervention across mental health, substance use, and crisis response settings.

Vision

To lead a transformative shift in behavioral health systems by integrating AI-driven and immersive innovations into real-world practice, enabling timely, precise, and equitable responses to mental health challenges at both individual and population levels.

Goals

  1. Innovate AI-Driven Behavioral Health Solutions
    Develop and validate advanced AI models and digital tools that enhance early identification of risk, personalize interventions, and improve behavioral health outcomes.
  2. Advance Immersive Training and Intervention Systems
    Design and implement virtual and augmented reality platforms to improve communication, crisis response, and de-escalation skills among practitioners and public safety professionals.
  3. Bridge Research and Practice
    Translate scientific findings into actionable, practitioner-facing tools and decision-support systems that can be deployed in clinical, community, and public safety settings.
  4. Promote Culturally Responsive and Human-Centered Design
    Ensure that all technologies and interventions are grounded in cultural understanding, equity, and the lived experiences of diverse populations.
  5. Build Interdisciplinary and Community Partnerships
    Foster collaboration across behavioral health, technology, public safety, and community sectors to maximize real-world impact and scalability.

Current Project

AI-Driven ReCoDe Virtual Reality Training for Policing

RecoDe VR for Policing Logo Fatal encounters between police and the public too often involve behavioral-health crises. Police are often the first to encounter community members experiencing acute mental-health crises (e.g., psychosis, severe anxiety) and/or substance-related crises (e.g., intoxication, stimulant-induced psychosis). In these high-stress encounters, miscommunication can escalate to use of force and civilian fatalities; national estimates indicate that roughly one in four deaths during law-enforcement contact involves a person with serious mental illness, and about one in five decedents show mental-illness symptoms or drug-induced disruptive behavior.

The gap we target is communication and de-escalation in moments complicated by psychosis, intoxication, or both. We propose an interdisciplinary, VR-integrated De-Escalation Training Collaborative to design a scalable training-and-assessment blueprint that strengthens officers’ crisis recognition, trauma-informed communication, and step-wise de-escalation to reduce harm to civilians. Using immersive, co-created scenarios—developed with law-enforcement trainers, behavioral-health clinicians, and community advisors—we will prototype a four-module curriculum covering (1) mental and behavioral health symptoms, (2) substance misuse and co-occurring conditions, (3) crisis-communication skills, and (4) stepwise de-escalation tactics, all practiced and assessed in an immersive VR environment. Pre-seed support will help toward convening the team, run scenario/design charrettes, align platform/analytics/privacy requirements, build an evaluation framework (behavioral competencies and rubrics), launch a project website, and cultivate agency/funder partners. The outcome would be a fundable package positioning our UGA team for competitive seed and external opportunities (e.g., NIJ, DoD, DOJ-COPS, SAMHSA, NSF, Foundations) and a subsequent multi-site pilot with rigorous mixed-methods evaluation—a path to scalable national dissemination of VR de-escalation training that reduces preventable deaths and harm across communities.

Affiliated Team

Hee Yun Lee

Hee Yun Lee, PhD, LICSW — Holland Distinguished Professor, School of Social Work (Lead), University of Georgia; behavioral-health content, training design, technology-driven intervention, and trial evaluation

Grace Ahn

Grace (Sun Joo) Ahn, PhD — Professor & Director, Center for Advanced Computer–Human Ecosystems, Grady College of Journalism and Mass Communication, University of Georgia; VR/immersive media and learning analytics and communication strategies in behavioral crisis

Dan Silk

Dan Silk, PhD — Associate Vice President for Public Safety, University of Georgia; Public-safety strategy and implementation, development of real-life cases, and GA-wide networking and connection

Jeff Clark

Jeffrey L. Clark, MPA — Chief, University of Georgia Police Department, University of Georgia; primary liaison to UGA PD and regional law-enforcement partners, and police site access and training implementation

Sarah K. S. Shannon

Sarah K. S. Shannon, PhD, MSW — Associate Professor of Sociology & Director, Criminal Justice Studies Program, University of Georgia; criminal-legal systems and evaluation, policy translation and dissemination, and training content development

Prashant Doshi

Prashant Doshi, PhD — UGA Foundation Distinguished Professor of Artificial Intelligence, Executive Director, Institute for Artificial Intelligence, University of Georgia; architect the AI back-end that powers the VR scenarios (dialog manager, intent recognition, and emotion/state inference) using NLP and decision-making models

Soobin Kim

Soobin Kim, PhD, MSW – Postdoctoral Research Associate, School of Social Work, University of Georgia; trial evaluation, policy translation and dissemination

Seunghyun Moon headshot

(Max) Seunghyun Moon, PhD Student


VR Engineering Support — Center for Advanced Computer–Human Ecosystems (https://www.ugavr.com), Grady College of Journalism and Mass Communication, University of Georgia; speech recognition, NLP integration for conversational branching, and performance analytics

Modules for Effective Communication on Behavioral Health

Module One – Introduction to Mental Health and Mental Illnesses
This course will cover the following:

  • The history of behavioral health approaches in the United States
  • Symptoms of common severe mental illnesses
  • The prevalence of mental illness
  • Co-occuring scenarios & officer response

Module Two – Mental Health and Substance Abuse Disorders
This course will cover the following:

  • The linkages between mental health and substance abuse
  • Commonly abused addictive substances in central Alabama
  • Co-occuring scenarios & officer response

Module Three – Communication Systems and Active Listening Techniques
This course will cover the following:

  • Communication as a system
  • Non-verbal communication and cues to help identify possible mental illness
  • Active listening skills

Module Four – De-escalation and Crisis Intervention Techniques
This course will cover the following:

  • Crisis intervention history and application
  • Scenarios from the Emmy Award-winning documentary, Ernie & Joe Crisis Cops
  • Local referral resources

Community Advisory Board Members

Coming Soon …

Previous Research Project

Increasing Mental Health Awareness in Our Community

REMIND

Tuscaloosa County’s nine police departments piloted the Remind mental health training curriculum in the spring of 2021. The original three-course curriculum has been expanded to include a new course dedicated to Communication & Active Listening. Completion of these four courses provides 6 CEUs for officers (the maximum online credits allowable through APOSTC for 2022). This unique and relevant training opportunity supports first responders with de-escalation strategies when dealing with mental illness and substance abuse disorders. Remind’s goal is to strengthen first responders’ confidence in providing comprehensive mental health assistance and effective crisis intervention with reduced stigma.

This exclusive training increases mental health literacy and substance abuse awareness while highlighting diversity, equity and inclusion, communication, and intersectionality within the community.

Over 150 Alabama first responders have completed the Remind Training since its launch. Survey results show a significant increase in the participating first responder’s knowledge of schizophrenia, generalized anxiety disorder, and major depressive disorder. Total scores of behavioral health knowledge increased after the training.

More information can be found at https://www.remindforle.com/about

Lee, H.Y. (2020-2023)
Principal Investigator. Mobile Mental Health Education, Awareness, and Learning (mHEAL) Initiative for Law Enforcement. $375,000.
Mental Health Awareness Training Grants, Department of Health and Human Services, Substance Abuse and Mental Health Services Administration (Grant Number: 1H79SM081438-01).

Join our commitment to well-being and positive social change.

Apply Today Make a Gift